JP4940891B2 - Product introduction robot, product introduction method and product introduction information processing program - Google Patents

Description

  The present invention relates to a product introduction robot, and more particularly, to a product introduction robot technique for introducing a product based on the intentions of each product manufacturer or dealer.

  As IT technology permeates home appliances and cameras, there are a wide variety of products developed by manufacturers of these electronic devices, and retailers that handle these huge numbers of products understand the specifications and characteristics of each product, There has been a limit to relying on the efforts of a single salesperson to accurately respond to customer requests and introduce products. Therefore, it is possible to divide the sales product territory and create a team of multiple sales staff to deal with the customer's request, or to have each product manufacturer dispatch a sales support helper to deal with it. It has begun.

  Since these methods are accompanied by an increase in labor costs, introduction of a product introduction robot that acts as a proxy or support for these product introduction tasks has been studied. For example, there is a customer service robot disclosed in Patent Document 1 below. Hereinafter, an outline of a necessary part of the product introduction robot disclosed in this document will be described with reference to FIG.

  The conventional product introduction robot in FIG. 11 is given from a product information database 91 that stores product information of products to be introduced, a request information acquisition unit 92 that acquires request information req from a customer c, and a request information acquisition unit 92. Product candidate selection means 93 for selecting a product candidate cg (generally a plurality of products) to be introduced by searching the product information database 91 with the search conditions to be extracted, and extracting the product candidate cg from the product information database 91 to obtain a comparison table An introduction presentation unit 94 having a comparison table creation unit 941 to be created, and the introduction presentation unit 94 presents a conventional product candidate item comparison table cgTBL ′ to the customer c. Further, the introduction presenting means 94 has an action executing means 942 and performs a product introduction operation act determined by obtaining the product identifier gid = x of the product to be introduced.

In the description of Patent Document 1, as the request information acquisition unit 92, the customer's utterance is recognized by speech to acquire the request information req, and the product presentation item comparison table cgTBL ′ by the introduction presentation unit 94 is, for example, the body of the robot This is performed by the image display means incorporated in. In addition, as the product introduction operation act, an example is described in which a robot performs an instruction action indicating an actual product.
JP 2005-293052 A

  In the product introduction robot of Patent Document 1, when a large number of product candidates cg appear, it cannot be narrowed down and presented to the customer c with priorities. For example, at a sales office of a mass retailer that handles digital cameras (hereinafter referred to as “digital cameras”), each of which has a large number of product series, a large number of candidates appear in the conventional product candidate item comparison table cgTBL ′, and the customer c is determined. It is expected that you will be in trouble. In addition, when a product introduction robot is dispatched from a product manufacturer, the product introduction robot wants to introduce the manufacturer's own products with priority if it is the same level. A certain priority product cannot be recommended. In addition, intentional introduction and presentation that emphasizes such priority products cannot be performed.

  As described above, in the conventional product introduction robot, it is not possible to recommend a priority product that is a product that the robot should preferentially introduce, and it is not possible to provide intentional introduction that emphasizes the priority product. There was a problem.

  In the conventional product introduction robot, the above-mentioned problem that the robot cannot recommend a priority product that is a product to be preferentially introduced and cannot perform intentional introduction that emphasizes the priority product. As shown in FIG. 1, a product introduction robot that introduces a product to the customer c in correspondence with the customer c at the product sales floor, and defines a priority product yg that is a product that the robot should preferentially introduce. Definition means 5, product information database 1 having a record made up of product identifier item gid and a plurality of product characteristic items gp1, 2,..., Request information acquisition means 2 for acquiring request information req of customer c, and the acquisition Based on the determined desired item gp-req, desired item determining means 3 for determining the product characteristic item gp1,2,... Related to the keyword k included in the determined desired information req as the desired item gp-req. , The product information Product candidate selection means 4 that searches the database 1 and selects a plurality of product candidates cg to be introduced, and when the priority product yg is included in the selected product candidate cg, the priority product yg is recommended. This is solved by a product introduction robot having intentional response means 6 for performing a corresponding output r-out.

  That is, in the product introduction robot of the present invention, the priority product yg is set in advance by the priority product definition means 5. Then, through the request information acquisition means 2 and the request item determination means 3, the request information req of the customer c is regarded as a request item gp-req to associate with the product information database 1, and the product information database 1 is searched by the product candidate selection means 4 to search for product candidates. Select cg. When this product candidate cg includes the priority product yg, the intentional response means 6 performs a corresponding output r-out that recommends the priority product yg, so that the customer c is more preferably impressed with the priority product yg, It becomes easy to lead to the purchase of this product.

  Further, as shown in FIG. 6, the problem is that the product characteristic item value val (yg, gpi) of the priority product yg exceeds the product characteristic item value val (cg, gpi) of another product candidate cg. The characteristic item gpi is determined as the advantageous item gp-y, and when the product characteristic item value of the priority product is lower than the product characteristic item value of the other product candidates, the product characteristic item is determined as the disadvantaged item gp-h. It is more effective by having the item determination unit 63 and the comparison table presentation unit 64 that introduces and presents the product candidate item comparison table cgTBL that includes the determined advantageous item gp-y and does not include the disadvantaged item gp-h. To be resolved.

  That is, the priority product definition means 5 defines the priority product yg to be preferentially introduced by the product introduction robot, and the advantage / disadvantage item determination unit 63 confirms whether or not the priority product yg is included in the product candidate cg. . If included, the advantage / disadvantage item determination unit 63 compares the value of the product characteristic item value val (gid, gpi) between the priority product yg and the other product candidate cg for each product characteristic item gpi and gives priority. An advantageous item gp-y, which is an advantageous item for the product yg, and a disadvantageous item gp-h, which is an unfavorable item, are determined. The comparison table presentation unit 64 creates a product candidate item comparison table cgTBL that includes the advantageous item gp-y and does not include the disadvantageous item gp-h, and introduces and presents it. Therefore, by such a commodity candidate item comparison table cgTBL, the visitor c is more impressed with the priority commodity yg and is likely to lead to the purchase of this commodity.

  Further, as shown in FIG. 9, the subject is a product introduction operation knowledge base 65 that defines the product introduction operations act1,2,... Of the robot in association with the product characteristic items gp1,2,. In order to allow the customer c to confirm the product characteristic item value val (cg, gpi) of the product candidate cg, the operation executing unit 66 that executes the product introduction operation act1,2,... Using the product introduction operation knowledge base 65. It is possible to solve the problem more effectively.

  That is, for example, if the advantageous item gp-y of the priority product yg is selected from the product introduction operation knowledge base 65 and the product introduction operation actx is performed, the advantage of the priority product yg can be further impressed. It becomes easy to lead to the purchase of this product.

  According to the product introduction robot of the present invention, it is possible to recommend a priority product that is a product to be preferentially introduced, and there is an effect that the product introduction according to the intention of the product manufacturer or the dealer can be performed.

  An embodiment of the product introduction robot of the present invention will be described with reference to FIGS. Note that the computer processing in the present invention is executed by a computer program developed on the main storage device of the computer. The computer program is provided in a CD- format including an auxiliary storage device connected to the computer. Provided by being stored in a recording medium such as a portable storage device such as a ROM or a main storage device or an auxiliary storage device of another computer connected to a network. It is loaded and executed on a storage device.

  A first embodiment of the product introduction robot of the present invention will be described with reference to FIGS. FIG. 1 is a configuration example of a product introduction robot of this embodiment. FIG. 2 is an operation flow of the product introduction robot. In the following, the operation of this product introduction robot will be described based on these drawings.

  First, in step S201 of FIG. 2, at the stage of preparation for opening a store, the robot owner interactively inputs the priority product yg, and the priority product definition means 5 of the robot stores it therein. In step S202, the robot owner interactively updates and sets the product information database 1. Here, once leaving FIG. 2, the product information database 1 will be described with reference to FIG.

  FIG. 3 shows an example of the data structure of the product information database 1. This example shows a database of digital camera products at a digital camera store. Each record includes a product identifier item gid and a plurality of product characteristic items gp1, 2,. For example, the record with product identifier gid = 01040 is a digital camera of model No. JX D50 from Canox, and today's price for this product is ¥ 36,750. In addition, the product characteristic item value val (gpi) is inferior in thinness and size, but is superior in weight, vividness of color, noise in dark areas, and image size. The sensitivity is moderate, the camera shake prevention function, and the liquid crystal size is large. In the product information database 1, such a record is prepared for all products.

  Returning to FIG. 2 again, in step S203, the general control unit (not shown) of the robot detects the customer c, approaches the customer c, activates the request information acquisition means 2, and acquires the request information req from the customer c. The request information acquisition means 2 acquires the request information req from the customer c, and the mode is performed by various known techniques such as voice input and dialogue using a display and a keyboard. In the present embodiment, the request information acquisition means 2 acquires Japanese spoken by the customer c as the request information req by voice input. In step S204, the request information acquisition unit 2 analyzes the acquired request information req, and extracts a keyword k corresponding to the keyword registered in the built-in request item determination table kgpTBL (see FIG. 4A). Here, the request information acquisition means 2 extracts a word by a morpheme analysis means using a well-known word dictionary (not shown), and extracts a keyword k of the topic by referring to a built-in keyword synonym dictionary (not shown). Here, the keyword k is a word representing the request of the visitor c to the camera, and is defined in the first column of the request item determination table kgpTBL shown in FIG. For example, when the request information req of the visitor c is “I want to go on a trip”, the request information acquisition means 2 extracts “travel” as a keyword k.

  Next, in step S205, the desired item determination means 3 determines the desired item gp-req and its weight coefficient from the desired information req and the keyword k with reference to the desired item determination table kgpTBL. For example, when the request item determination table kgpTBL is registered as shown in FIG. 4 (a), the request item gp-req for the keyword `` athletic meet '' is thinness, size, camera shake prevention, and telephoto, and each has a weight coefficient. Is set.

  Next, in step S206, the product candidate selection means 4 searches the product information database 1 based on the desired item gp-req, calculates the desired score reqp for all product records, and selects the records for which the desired score reqp exceeds a predetermined threshold value. Extract as candidate cg records. The request score reqp of each item is, for example, 1 item for which the product characteristic item value val (gpi) in the product information database 1 in FIG. 3 is superior (○), 0.5 for the middle (△), and × for the inferior (×) 0 points are multiplied by a weighting factor. In FIG. 4 (b), in the above example, the product identifier gid = 01040 is 0 points thin, size 0 points, camera shake prevention 50 points, and the product identifier gid = 14011 is 20 points thin, size 30 points. In the product identifier gid = 02011, there are 0 points for thinness, 30 points for size, 0 points for preventing camera shake, and the like. In this way, the desired score reqp of the digital camera is calculated as the sum of all items, and if this exceeds a predetermined threshold value, for example, 75 points, the digital camera is selected as a product candidate cg. FIG. 4 (b) shows that the above three models have been selected as product candidates cg as a result.

  Next, the product candidate selection means 4 determines whether or not the product candidate cg has been extracted in step S207, and if affirmative, the process proceeds to the next step S208, but if negative, the process returns to step S203, and new request information req Try to get.

  When the product candidate cg can be extracted, in step S208, the dialog generation unit 62 of the intention response unit 6 extracts the product identifier gid = x of the priority product yg from the priority product definition unit 5, and the product identifier of the product candidate cg record Check if x is included in gid. In step S209, it is determined whether the priority product yg is included in the product candidate cg. When the priority product yg is included, the dialog generation unit 62 searches the dialog knowledge base 61 using the keyword k, the priority product yg, and the product candidate cg in step S210, selects an appropriate utterance configuration program, The utterance is the corresponding output (voice) r-out. This will be described later using a specific example. Then, the request information acquisition means 2 determines in step S211 whether or not the customer c has decided on the purchased product, and if there is still a request, returns to step S203 and continues the dialogue with the customer c. Further, when the priority product yg is not included in the product candidate cg, the dialog generation unit 62 performs an utterance for sequentially explaining the product candidate cg.

A specific example in which the product introduction robot makes an output (voice) r-out utterance to the customer c by the operation of FIG. 2 will be described below. In this example, it is assumed that the product introduction robot has a product identifier gid = 14 *** (* is a wild card), that is, all products manufactured by Maxxxx as priority products yg.
<< Example of utterance >>
Customer: Hello. I would like a digital camera. Robot utterance 1: Let me show you the best selling. The Maxxxx company has a good reputation for its reflection. If you match this with the A mark, it is easy for anyone in auto mode ...
Customer: I want to go on a trip.
Robot utterance 2: Light is recommended for travel. If it ’s light,
These three models are recommended.

If you choose a large LCD, you can check it on the spot.
Customer: I've heard that it is better to prevent camera shake.
Robot utterance 31: It's a function that doesn't shake even in dark places.
Robot utterance 32: Although Camerax is a big proponent of camera shake prevention, Maxxxx has increased the shutter speed to prevent camera shake in the same way.
Customer: Sounds heavy.
Robot utterance 4: Take it with you. I don't feel it because it fits my hand.

And if it ’s a difference of 10 grams, if you take a sip of a plastic bottle, it ’s the same, so it ’s not the amount you care about.
Robot utterance 5: If you choose Maxxxx or Canoxx. The commercial is flashy at Canoxx.

Maxxxx makes it easy to get beautiful shots, switch it on and move quickly so you never miss a chance.

Maxxxx is now doubling the points, so you can add a lot of memory. How is it.
<< Explanation >>
Robot utterance 1: Request information acquisition means 2 could not extract a specific keyword k, so the keyword k = (default) is output, and the request item determination means 3 is the request item gp-req as the request item in FIG. 4 (a). In the judgment table kgpTBL, the item identifier item “Selling Goods” was selected. The product candidate selection means 4 searches the product information database 1 for this, and the three models shown in FIG. 4 (b) are selected as product candidates cg. Since the product identifier gid = 14 *** of the priority product yg is set, the dialog generation unit 62 searches the dialog knowledge base 61 shown in FIG. The program apeal ("", 01, 14011) was selected. The dialog generation unit 62 executes this program, and the utterance is configured.

  Robot utterance 2: Similarly, the utterance composition program weight ("travel", 61, "") is selected from the dialogue knowledge base 61 by the keyword k = travel, utterance subject = weight, and the above utterance is composed by executing this program. It was done.

  Robotic utterance 31: Similarly, the utterance composition program bure ("shake", 11, "") is selected from the dialogue knowledge base 61 by the keyword k = camera shake, utterance subject = camera shake prevention, and the above utterance is composed by executing this program. It was done.

  Robot utterance 32: Similarly, the utterance composition program bure ("shake", 11, "") is selected from the dialogue knowledge base 61 by the keyword k = hand shake, utterance subject = shutter speed, and the above utterance is composed by executing this program. It was done.

  Robot utterance 4: Similarly, the utterance composition program appeal ("Weight", 03,14011) is selected from the dialogue knowledge base 61 by the keyword k = weight, utterance subject = priority product appeal, and the above is executed by executing this program. The utterance was composed.

  Robot utterance 5: Similarly, the utterance composition program appeal ("", 02,14011) is selected from the dialogue knowledge base 61 by the keyword k = (default), utterance subject = priority product appeal, and the above utterance is executed by executing this program. Was configured.

  The product introduction robot of this embodiment has the configuration shown in FIG. 6, and operates according to the operation flow shown in FIG. Note that the configuration example of the product information database 1 in FIG. 3 and the desired item determination table kgpTBL in FIG. 4 are the same as those in the first embodiment. The block configuration of FIG. 6 is the same as that of FIG. 1 of the first embodiment, except that the internal configuration of the intentional correspondence means 6 is different. Therefore, this part will be mainly described. The intentional handling means 6 of this embodiment has an advantage / disadvantage item determination unit 63 and a comparison table presentation unit 64, and differs from the first example in that the product candidate item comparison table cgTBL is presented as the corresponding output r-out. .

  The operation of the product introduction robot will be described with reference to the operation flow shown in FIG. Again, steps S701 to S707 are the same as steps S201 to S207 already described in FIG. In step S708, the advantage / disadvantage item determination unit 63 of the intentional response unit 6 extracts the product identifier gid = x of the priority product yg from the priority product definition unit 5, and checks whether x is included in the product identifier gid of the product candidate cg record. In step S709, it is determined whether the priority product yg is included in the product candidate cg.

  When the priority product yg is included in the product candidate cg, the process proceeds to step S710, where the advantageous / disadvantaged item determining unit 63 determines the advantageous item gp-y and the disadvantageous item gp-h, and activates the comparison table presenting unit 64. . In step S711, the comparison table presentation unit 64 creates a product candidate item comparison table cgTBL that does not include the disadvantageous item gp-h but includes the advantageous item gp-y. FIG. 8 shows an example of creating the product candidate item comparison table cgTBL. As shown in FIG. 8 (a), when the product identifier gid = 14 *** of the priority product yg is set, the advantage / disadvantage item determination unit 63 searches the product information database 1 of FIG. The product characteristic item gp = (gp4, gp5, gp6, gp7, gp9, gp11, gp12, gp13, ...) of the advantageous item gp-y and the product characteristic item gp = of the disadvantageous item gp-h as shown in (b) (Gp8, gp10, gp14, ...) As a result, the comparison table presenting unit 64 presents a commodity candidate item comparison table cgTBL as shown in FIG. 8 (c).

  If the priority product yg is not included in the product candidate cg in the determination in step S709, the process proceeds to step S712, and the comparison table presenting unit 64 creates the product candidate item comparison table cgTBL including all items as in the past. . In step S713, the intention corresponding means 6 presents the product candidate item comparison table cgTBL created in step S711 or S712 to the customer c.

  The product introduction robot of the present embodiment has the configuration shown in FIG. 9, the configuration example of the product information database 1 in FIG. 3, the request item determination table kgpTBL in FIG. 4, and the configuration example of the dialog knowledge base 61 in FIG. It is common. The block configuration of FIG. 9 is the same as that of FIG. 1 of the first embodiment except that a product introduction operation knowledge base 65 and an operation execution unit 66 are added inside the intentional correspondence means 6. Therefore, this part will be mainly described.

  As in the first embodiment, in the product introduction robot, the intentional response means 6 speaks with a corresponding output (voice) r-out in response to the request information req of the customer c. In addition, in the product introduction robot of the present embodiment, the intentional response means 6 performs various product introduction actions act1, 2,. The operation program is stored in the product introduction operation knowledge base 65 as a program group. FIG. 10 shows a configuration example of the product introduction operation knowledge base 65.

  As shown in FIG. 10, in order for the product introduction robot to perform the product introduction operation, functional requirements that the robot should have are necessary. For example, in order to perform the product introduction operation act0311 in which the robot measures and presents the weight of the product, the robot must have a hand-held function and a weight measurement function. In general, robots that have all these functional requirements are large and expensive, so practically, several types of product introduction robots with different shapes and functions have been created. .. Are selected from the product introduction operation knowledge base 65 within the range of functional requirements possessed by.

In such a product introduction robot, in addition to the dialogue as in the first embodiment, the corresponding output r-out with the following product introduction actions act1, 2,.
<< Example of supported output r-out >>
Customer: Sounds heavy.
Robot utterance 6: Weigh this camera.

Here we see the weight.
Customer: Can you take beautiful pictures?
Robot utterance 7: Let's take a picture.

    Try to shoot while moving like this, and display the result on this LCD.

First is the Cannoxx camera. There is no camera shake, but when you zoom in, the noise becomes so noticeable.

    Next time I will shoot Maxxxx. The noise will not stand out even if you zoom in.

      This is recommended for taking beautiful pictures.

  In the above robot utterance 6, the dialogue generator 62 executes the utterance composition program action ("weight", 88,14011) from the dialogue knowledge base 61 of FIG. 5 with the keyword k = weight and utterance subject = actual measurement. This made up the utterance. Then, an operation trigger trig from the program action (“weight”, 88, 14011) is given to the operation execution unit 66, and the operation execution unit 66 selects the product introduction operation act0311 from the product introduction operation knowledge base 65 based on the parameters. While the operation executing unit 66 executes this, the corresponding output (voice) r-out by the dialogue generating unit 62 is uttered in synchronization.

  In the robot utterance 7, the dialogue generator 62 executes the utterance composition program action (“clean”, 88, 14011) from the dialogue knowledge base 61 of FIG. 5 with the keyword k = clean and the utterance subject = actual measurement. This made up the utterance. Then, an operation trigger trig from the program action (“clean”, 88, 14011) is given to the operation execution unit 66, and the operation execution unit 66 selects the product introduction operation act0612 from the product introduction operation knowledge base 65 according to the parameters, While the action execution unit 66 executes this, the corresponding output (voice) r-out by the dialogue generation unit 62 is uttered in synchronization.

  In both the first, second, and third embodiments, it is easy to provide a management computer that communicates with the product introduction robot and records the corresponding operation of the product introduction robot. In addition, such a management computer records the contribution including the customer service frequency and sales of each product introduction robot, the money is paid to the product introduction robot owner according to the contribution, and the set priority product Each time, a business model is devised, such as payment of money from the priority product manufacturer to the store according to the degree of contribution, but these are considered to be design matters at the time of implementation.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1) A product introduction robot that introduces products to customers in response to visitors at the product sales floor.
Priority product definition means for defining a priority product that is a product to be preferentially introduced by the robot;
A product information database having a record composed of a product identifier item and a plurality of product characteristic items;
Request information acquisition means for acquiring the request information of the visitor;
Request item determination means for determining the product characteristic item related to the keyword included in the acquired request information as a request item;
Product candidate selection means for searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
When the selected product candidate includes the priority product, an intentional response unit that performs a corresponding output to recommend the priority product;
A product introduction robot characterized by comprising:
(Appendix 2) The intentional response means further includes:
When the product characteristic item value of the priority product exceeds the product characteristic item value of another product candidate, the product characteristic item is determined as an advantageous item, and the product characteristic item value of the priority product is a product characteristic item of another product candidate An advantageous / disadvantaged item determination unit that determines the product characteristic item as an adverse item when the value is below the value;
A comparison table presenting section for introducing and presenting product candidate item comparison tables that include the determined advantageous items and do not include disadvantageous items;
The product introduction robot according to supplementary note 1, characterized by comprising:
(Appendix 3) The intentional response means further includes:
Product introduction operation knowledge base that defines the product introduction operation of the robot in association with the product characteristic item,
In order to make the customer confirm the product characteristic item value of the selected product candidate, using the product introduction operation knowledge base, an operation execution unit that executes the product introduction operation;
The product introduction robot according to appendix 1 or 2, characterized by comprising:
(Appendix 4) A product introduction method for a product introduction robot for introducing a product to a visitor in response to a visitor at a product sales floor,
A product information database having a record made up of product identifier items and a plurality of product characteristic items,
A priority product definition step for defining a priority product that is a product to be preferentially introduced by the robot;
A request information acquisition step for acquiring the request information of the visitor;
A request item determination step of determining, as a request item, the product characteristic item related to the keyword included in the acquired request information;
A product candidate selection step of searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
When the selected product candidate includes the priority product, an intentional response step of performing a corresponding output that recommends the priority product;
A product introduction method characterized by comprising:
(Appendix 5) A product introduction information processing program for causing a computer to execute a product introduction robot for introducing a product to a visitor in response to a visitor at a product department,
A product information database having a record made up of product identifier items and a plurality of product characteristic items,
A priority product definition step for defining a priority product that is a product to be preferentially introduced by the robot;
A request information acquisition step for acquiring the request information of the visitor;
A request item determination step of determining, as a request item, the product characteristic item related to the keyword included in the acquired request information;
A product candidate selection step of searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
When the selected product candidate includes the priority product, an intentional response step of performing a corresponding output that recommends the priority product;
A product introduction information processing program characterized by causing a computer to execute.
(Appendix 6) A computer-readable recording medium recording a product introduction information processing program for causing a computer to execute a product introduction robot for introducing a product to a visitor corresponding to a visitor at a product sales floor,
A product information database having a record made up of product identifier items and a plurality of product characteristic items,
A priority product definition step for defining a priority product that is a product to be preferentially introduced by the robot;
A request information acquisition step for acquiring the request information of the visitor;
A request item determination step of determining, as a request item, the product characteristic item related to the keyword included in the acquired request information;
A product candidate selection step of searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
When the selected product candidate includes the priority product, an intentional response step of performing a corresponding output that recommends the priority product;
A computer-readable recording medium on which a product introduction information processing program is recorded.

The product introduction robot of the present invention allows a product introduction that recommends a priority product to the robot by designating a priority product that is a product to be preferentially introduced. There is a possibility that it will be used by substituting or supporting a manufacturer's dispatch sales helper business.

Block configuration example of the product introduction robot of the first embodiment of the present invention Operation flow diagram of the product introduction robot of the first embodiment Data structure example of product information database 1 Example data of request item judgment table kgpTBL and product candidate cg Data structure example of Dialogue Knowledge Base 61 Block configuration example of the product introduction robot of the second embodiment of the present invention Operation flow diagram of the product introduction robot of the second embodiment Product candidate item comparison table cgTBL creation example Block configuration example of the product introduction robot of the third embodiment of the present invention Data structure example of product introduction behavior knowledge base 65 Block configuration example of a conventional product introduction robot

Explanation of symbols

1 Product information database
2 Request information acquisition means
3 Desired item determination means Product candidate selection means Prioritized product definition means Intentional response means Dialogue knowledge base dialog generation part Advantage / Disadvantage item judgment part Comparison table presentation part Product introduction action Knowledge base action execution part
req Request information
k keyword
gp-req requested item
gid Product identifier field
gp1,2,… Product characteristic item
yg priority product
cg Product candidate
r-out compatible output
gp-y advantage
gp-h disadvantageous item
cgTBL Product Candidate Item Comparison Table
act1,2,… Product introduction action

Claims (3)

A product introduction robot having a computer that corresponds to the visitor under the trade department to introduce a product to the visitor,
The computer
Priority product definition means for storing in advance information on priority products that the robot should preferentially introduce;
A product information database that is stored and saved in advance association with item identifier and a plurality of product characteristics fields for identifying the goods,
Request information acquisition means for acquiring the request information of the visitor;
As requested item the item characteristic items associated with keywords contained in the acquired the requested information, and request item determination means makes determination by using the request item determination table weighting coefficient of the product characteristics item is set,
Based on the determined desired item, product candidate selection means for searching the product information database and selecting a plurality of product candidates to be introduced;
A product introduction operation knowledge database in which the product characteristic items and the product introduction operation data of the robot are associated and stored in advance;
When the selected product candidate includes the priority product, an operation for causing the customer to introduce a product by executing a related product introduction operation action from the product characteristic item stored in the product introduction operation knowledge database. An action execution unit for displaying
A product introduction robot characterized by comprising: A product introduction method of a product introduction robot having a computer for introducing a product to the visitor corresponding to the visitor at a product sales floor,
A product information database that stores and saves product identifiers that identify products and a plurality of product characteristic items in advance,
By the computer,
A priority product definition step in which information of priority products to be preferentially introduced by the robot is stored in advance;
A request information acquisition step for acquiring the request information of the visitor;
A request item determination step for determining the product characteristic item related to the keyword registered in the request item determination table included in the acquired request information as a request item;
A product candidate selection step of searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
A product introduction operation related to the product characteristic item stored in the product introduction operation knowledge database by searching a product introduction operation knowledge database in which the product characteristic item and the product introduction operation data of the robot are stored in association with each other are stored. An action execution step for executing an action and displaying an action for introducing a product to the visitor;
A product introduction method characterized by including: In the product introduction robot computer that introduces the product to the visitor corresponding to the visitor at the product sales floor,
A product information database that stores and saves product identifiers that identify products and a plurality of product characteristic items in advance,
A priority product definition function that stores in advance information on priority products that the robot should preferentially introduce,
A request information acquisition function for acquiring the request information of the visitor;
A request item determination function for determining the product characteristic item associated with the keyword registered in the request item determination table included in the acquired request information as a request item;
A product candidate selection function for searching the product information database and selecting a plurality of product candidates to be introduced based on the determined desired items;
The product introduction operation knowledge is searched by searching a product introduction operation knowledge database in which the product characteristic items and the product introduction operation data of the robot are stored in association with each other.
An operation execution function for causing a product introduction operation action to be executed from the product characteristic item stored in a database and displaying an operation for introducing a product to the customer;
Product introduction information processing program to realize.

Images